OVERALL DESCRIPTION: (Adapted from the Investigator's Abstract) This renewal application proposes the development of a computational NMR facility. The long term objective of this program is to develop computational methods for determining structures and motions of biological macromolecules in noncrystalline environments. The specific aims are: 1) Quantitative uses of 2D NMR structural data, especially Nuclear Overhauser Effect (NOE) experiments and coupling constants; 2) Development of linear prediction processing algorithms to yield chemical shift, linewidth, and intensity data directly; 3) Extending computer-assisted spectral interpretation to include peak identification, spin system assignment, and sequential assignment. These methods will use computer graphics, structural databases, and pattern recognition; 4) Developing a complete package of algorithms for constrained optimization of distance geometry (DG), relaxation matrix methods, and molecular dynamics (MD); 5) Analysis of direct relaxation data for studies of molecular motions. It is emphasized that in-house determination of accurate data is an essential component of this proposal which describes experimental applications that will make use of this facility. The project requires access to high field spectrometers and considerable computational resources. Because of the enormous potential of the method and the significant effort required to bring full project to maturity, it has been requested as a resource-related research project, coordinated with the Research Resource at the Computer Graphics Laboratory (CGL) of the University of California.